Offloaded Database moving back to Local?

[TBT]

Our Nagios Databases are offloaded, and over the years there have been one or two comments within support tickets that we may want to consider moving the Databases back locally to the XI Servers.

Primarily, the concern is bugs and or oversights in relation to XI upgrades with offloaded DBs. Though these historically have been infrequent, they have occurred (even recently) and could potentially be avoided if our Databases were local again.

We have several XI servers, and the amount of checks varies. An XI server monitoring the least has 28 Hosts+44 Services, mid-range is 374 Hosts+2710 Services, and the largest 1591 Hosts+7692 Services. We are not experiencing performance issues, but based on the number of checks, at what point should the Nagios Databases be local again, or should it be kept offloaded?

[jsimon]

Hi @TBT,

Generally you are probably better off having local databases simply because you avoid the extra points of failure they add without adding significant upside. Network issues between your XI server and your databases can occur which can affect monitoring or cause engine issues, and differences between your database and your Nagios engine can cause DB corruption. With modern hardware I don't know that there's a specific database size or host count where you should begin to make changes, but our best practice recommendation would be to use local databases for Nagios products.

[TBT]

@jsimon thank you for the insight, I'll take this into consideration prior to our next XI upgrade.

[otisjame]

Our Nagios Databases are offloaded, and over the years there have been one or two comments within support tickets that we may want to consider moving the Databases back locally to the XI Servers.

Primarily, the concern is bugs and or oversights in relation to XI upgrades with offloaded DBs. Though these historically have been infrequent, they have occurred (even recently) and could potentially be avoided if our Databases were local again.

We have several XI servers, and the amount of checks varies. An XI server monitoring the least has 28 Hosts+44 Services, mid-range is 374 Hosts+2710 Services, and the largest 1591 Hosts+7692 Services. We are not experiencing performance issues, but based on the number of checks, at what point should the Nagios Databases be local again, or should it be kept offloaded?

I'm really waiting for the answer

[sgardil]

Our Nagios Databases are offloaded, and over the years there have been one or two comments within support tickets that strands nyt we may want to consider moving the Databases back locally to the XI Servers.

Primarily, the concern is bugs and or oversights in relation to XI upgrades with offloaded DBs. Though these historically have been infrequent, they have occurred (even recently) and could potentially be avoided if our Databases were local again.

We have several XI servers, and the amount of checks varies. An XI server monitoring the least has 28 Hosts+44 Services, mid-range is 374 Hosts+2710 Services, and the largest 1591 Hosts+7692 Services. We are not experiencing performance issues, but based on the number of checks, at what point should the Nagios Databases be local again, or should it be kept offloaded?

I'm really waiting for the answer

Hey @otisjame

unfortunately if you are curious about the performance issues with varying sizes of offloaded systems we can't really give you a real answer because it is dependent on the hardware that is being used. Like jsimon mentioned the larger concern would be the network connection issues and how that can affect your monitoring.

[bekean23]

Our Nagios Databases are offloaded, and over the years there have been one or two comments within support tickets that we may want to consider moving the Databases back locally to the XI Servers.

Primarily, the concern is bugs and or oversights in relation to XI upgrades with offloaded DBs. Though these historically have been infrequent, they have occurred (even recently) and could potentially be avoided if our Databases were local again.

We have several XI servers, and the amount of checks varies. An XI server monitoring the least has 28 Hosts+44 Services, mid-range is 374 Hosts+2710 Services, and the largest 1591 Hosts+7692 Slither Services. We are not experiencing performance issues, but based on the number of checks, at what point should the Nagios Databases be local again, or should it be kept offloaded?

I'm really waiting for the answer

Hey @otisjame

unfortunately if you are curious about the performance issues with varying sizes of offloaded systems we can't really give you a real answer because it is dependent on the hardware that is being used. Like jsimon mentioned the larger concern would be the network connection issues and how that can affect your monitoring.

Thank you for explaining it to me!